Multichannel, Battery-less, and Wireless Recorder for Neuronal Activity Monitoring in Swine

Introduction: In 2015, 1.2% of the total US population had active epilepsy [1]. While epilepsy can be caused by a variety of conditions such as stroke, brain tumor or traumatic brain injury, more often, the cause is unknown. Electrocorticogram (ECoG) is frequently used to identify the origination of seizures. But this recording involves wires that protrude from the skull, rendering the patients bed-bound, and exposing them to increased risk of infection and hemorrhage. To address these issues, several implantable ECoG systems have been designed. However, they often require a power supply that can generate heat, possibly causing tissue damage. Our group has been recognized for introducing and developing a battery-less, multichannel, wireless neurosensing system (mWiNS) [2]. The operation of this system is based on radiofrequency (RF) backscattering and optical communications for multiplexing capabilities. Benchtop and in vivo evaluation in rodents of the system has been performed in the past [3],[4]. However, we have not been able to fully implant the multichannel system in vivo because of its large footprint relative to the size of rodent skull and brain. This study addresses these issues by introducing two new approaches. First, the multichannel system is reduced in size with optimal packaging for chronic in vivo use. Second, for the first time, we perform full implantation in a large, regulated animal model (swine) and successfully record evoked neural potentials. As such, we have developed a process for evaluating the system in chronic epileptic pigs in the future.

Materials and

Methods: WiNS is comprised of an implanted neuropotential recorder and an exterior interrogator (Fig. 1A). The implanted sensor is connected to neural probes, and is coated with a PDMS polymer for biocompatibility. For recording, the interrogator sends a carrier excitation signal that is mixed with the neural signals captured and multiplexed by the recorder. This modulated/multiplexed signal is then sent back to the interrogator for posterior demodulation and demultiplexing of the neural signals. The experimental procedures were approved and performed in swine in compliance with the Institutional Animal Care and Use Committee (IACUC) at Florida International University (Approval No. 23–023). The data presented in this paper were obtained from measurements on a female swine. For this experiment, a Ag/AgCl probe was fabricated in our lab for implantation with an electrochemically coated silver wire using chloride [3]. For multichannel data, an ECoG array (8 channels, FDA approved) from AD-Tech Medical Instruments Corporation was employed. A craniotomy was performed to place the ECoG array on a cortex (Fig. 2A), targeting the brain region of interest. The Ag/AgCl probe served as the electrical ground connected to a screw in the bone. The recordings were performed under complete anesthesia. Small needle electrodes (gauge 26, 1'' length) were inserted in the pig hindlimb to provide electrical pulses. The latter had an electric current amplitude of 5-35 mA at 2 Hz with a negative pulse form and a width of 200 μs, generated using an isolated pulse stimulator (AM Systems Model 4100).

Results, Conclusions, and Discussions:

Results:
The implant and interrogator were designed and fabricated using multilayers and 3D additive manufacturing technology, leading to a 52% smaller footprint than previous designs (Fig. 1B) [4]. The communication link between the implant and interrogator antennas was improved, demonstrating better performance over previous designs (Fig. 1C). The capability of recording somatosensory evoked potentials (SSEPs) was evaluated in swine. The recorded raw data from the hindlimb stimulation was filtered using a 60Hz notch filter and a band-pass filtered across 0.1-100Hz. Each recording (2min) was segmented from -50 to 250 ms, referenced to the time when the stimulation occurred. Subsequently, SSEPs were extracted by averaging across the signal segments. A comparison between a control signal and SSEP was performed for single channel recordings (Fig. 2B). The control signal refers to the average of the recordings with no stimulation. Additionally, SSEPs from 8-channel activation are also analyzed by digitizing the demodulated dataset at a sampling rate of 500 Hz per channel. Recovered SSEPs are shown in Fig. 2C with respect to a common reference.

Conclusions:
We developed a novel 3D additive sensor package that is multichannel, wireless, battery-less and matched to high impedance commercial electrodes. The implant size was reduced by 52% as compared to previous ones by our team, and sensitivity was also improved. For the first time, SSEPs were recorded from swine with our mWiNs. Future work will focus on epilepsy recordings in pigs.

Discussions:
These results demonstrate the potential of mWiNS to significantly impact future neuroscience research. The battery-less and wireless recorded data could be employed to analyze the patient’s well-being in real-time, offering a diagnosis and treatment for several neurological disorders such as epilepsy.

References:
[1] Zack MM, Kobau R, Morb Mortal Wkly Rep 2017;66:821–825.
[2] W. Chen, C. Lee, A. Kiourti, and J.L. Volakis, IEEE JERM, vol. 2, no. 4, pp. 262-269, 2018.
[3] C. Moncion, L. Balachandar, S. Bojja-Venkatakrishnan, J. J. Riera, and J. L. Volakis, IEEE JERM, vol. 3, no. 3, pp. 199–205, 2019.
[4] C. Moncion, L. Balachandar, S. B. Venkatakrishnan, J. L. Volakis, and J. Riera Diaz, Biosens Bioelectron. vol. 213, 114455, 2022.

Acknowledgements (Optional): This work was supported by National Science Foundation award 2052764 and the Wallace Coulter Foundation. The device 3D fabrication was done by Nano Dimension Co.